Chapter II. Phospholipase Cβ1 potentiates glucose-stimulated insulin secretion
2.4. Conclusion and Discussion
35
metabolic syndrome known to cause an increased functional response in pancreatic β-cells. HFD induced a significant, progressive increase in body weight in both Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice, although no significant differences in weight gain were observed between groups (Figure 2.7A).
In addition, islet morphology and β-cell proportion were not distinguishable between the two groups (Figure 2.7B). Next, we performed the glucose tolerance test to analyze glucose homeostasis in Plcb1f/f; Pdx1-CreERt2 mice during HFD intake. Plcb1f/f; Pdx1-CreERt2 mice receiving the HFD exhibited marked defects in glucose clearance in the blood relative to Plcb1f/f mice (Figure 2.7C), with no change in insulin sensitivity (Figure 2.9B) or total insulin content (Figure 2.7D) between groups. This pronounced deficiency in glucose tolerance was correlated with reduced plasma insulin levels, which led to elevated glucose levels in Plcb1f/f; Pdx1-CreERt2 mice (Figure 2.7E). Correspondingly, the GSIS assay in cultured islets from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice fed a HFD showed that the stimulatory effect of high glucose was lower in Plcb1f/f; Pdx1-CreERt2 islets than in those of Plcb1f/f islets (Figure 2.7F). Thus, PLCβ1 is necessary to maintain normal insulin secretion from β-cells and this mechanism is more prominent in a HFD-induced state.
36
cells of adult mice led to glucose intolerance due to decreased insulin secretion without altering insulin content or affecting β-cell architecture. In addition, the Pdx1 promoter showed restricted Cre activity primarily within the hypothalamus and Cre-mediated recombination in hypothalamus region could be involved in regulation of food intake or body weight [39]. Although PLCβ1 expression was decreased in hypothalamus, Plcb1f/f; Pdx1-CreERt2 mice showed no alteration in food intake or body weight (Figure 2.12A, B). The impaired insulin secretion in Plcb1f/f; Pdx1-CreERt2 mice was attributed to an intrinsic defect in β-cells, as isolated islets from these mice exhibited lower GSIS than those of Plcb1f/f mice. Our data showed that PLCβ1 exerted its effects exclusively via GSIS augmentation, making it important to determine the mechanism through which glucose induces PLCβ1-mediated insulin secretion. Glucose can stimulate the production of inositol phosphates in pancreatic islets, and PLCβ is a class of enzymes involved in phosphoinositide hydrolysis. Therefore, glucose is likely to stimulate PLC activity, leading to the accumulation of inositol trisphosphate and inositol tetrakisphosphate [40, 41]. Despite a lack of evidence supporting the involvement of mediators, glucose may couple the PLCβ- mediated production of inositol phosphates and insulin secretion [17, 42].
Here, we confirmed that PLCβ1 contributed to 5-HT-, VAP-, and KP-dependent insulin secretion.
In pancreatic islets, innervation and vascularization are important mechanisms in the regulation of insulin secretion. Notably, modulators such as 5-HT, released from pancreatic nerve endings, and vasopressin or kisspeptin, released from blood capillaries, activate PLCβ-coupled G proteins and augment insulin release under high glucose conditions. Moreover, β-cells co-secrete various substances, including 5-HT, with insulin under high glucose stimulation. These mediators from blood, nerve, and insulin vesicles could activate GPCR/PLCβ1 signaling in a paracrine and autocrine manner, leading to increased insulin secretion [7-10]. In particular, 5-HT-dependent insulin secretion is more prominent during pregnancy. Pregnancy results in increased β-cell proliferation and GSIS to compensate for the enhanced insulin demand through 5-HT signaling [37, 43]. Indeed, maternal lactogens significantly induce the synthesis and secretion of 5-HT, which activates the Htr2b receptor on β-cells to augment GSIS [44]. In male adult Plcb1f/f; Pdx1-CreERt2 mice, there was no remarkable change in β-cell proliferation and β-cell turnover (Figure 2.13A, B). However, PLCβ1 signaling that regulates GSIS through serotonergic pathways might also be active during pregnancy, and PLCβ1 could be the genetic cause of gestational diabetes, which has great potential for developing into type 2 diabetes mellitus [45].
GPCRs can bind specific PLCβs and modulate intracellular signaling by binding distinct partners, such as A-kinase anchoring proteins and postsynaptic density disc-large ZO-1 proteins [46, 47].
However, further studies are needed to clarify which binding partners are engaged in GqPCR/PLCβ1 signaling.
37
In this study, alterations to specific ligands, such as CCK, OXO-M, and OA, did not produce any differences in insulin secretion between pancreatic islets from Plcb1f/f; Pdx1-CreERt2 and Plcb1f/f mice.
One possible explanation for this distinction is functional compensation. Although there was no compensatory upregulation of PLCβ isozymes (Figure 2.14A, B), due to a shared primary structure and similar enzymatic activity, other still expressed PLCβ isozymes might be engaged with these GqPCRs.
In addition, this finding indicates the involvement of G protein families other than Gq, such as the Gs
family [48, 49]. Indeed, treatment of pancreatic islets with carbamylcholine, a muscarinic receptor agonist, and CCK results increased cAMP levels, which can enhance insulin secretion, indicating that they could control insulin secretion in a PLCβ1-independent manner [50]. In fact, CCK1 receptors have been reported to couple to both Gq and Gs. The CCK1-Gs pathway has been shown to regulate insulin secretion under high glucose conditions, while CCK1-Gq signaling is able to regulate insulin secretion under low glucose conditions [51].
The glucose-stimulated [Ca2+]i increase in β-cells requires Ca2+ influx via membrane depolarization and activation of L-type calcium channels. Besides IP3-mediated Ca2+ efflux from the endoplasmic reticulum (ER), PLCβ can contribute to further [Ca2+]i increases via its function as an important upstream signaling molecule for activation of store-operated channels (SOCs) and transient receptor potential channels (TRPCs) [52]. Depletion of ER Ca2+ stores via the IP3 receptor leads to activation of SOCs on the plasma membrane. Local reduction of PIP2 and increased DAG directly facilitate activation of TRPC3 [53], with PKC also activating nonselective cation channels, increasing [Ca2+]i [18]. However, despite these observations, further studies are required about a direct association between the PLCβ pathway and activation of Ca2+ channels.
Normal pancreatic β-cells have the capacity to secrete adequate levels of insulin necessary to compensate for increased insulin demands under conditions of HFD-associated insulin resistance [54].
Indeed, in islets isolated from mice fed HFD, pancreatic islets were expanded and secreted insulin levels were augmented, about 0.5 ng/µg protein/h, as compared to those fed normal chow diet (NCD), about 0.1 ng/µg protein/h. Also, basal insulin levels were increased in HFD-fed mice, around 0.5 ng/ml, as compared to NCD-fed mice, around 0.2 ng/ml. In Plcb1f/f; Pdx1-CreERt2 mice, HFD-feeding induced defects in glucose homeostasis. Similar to the role of PLCβ1 in regulating insulin secretion in normal Plcb1f/f; Pdx1-CreERt2 mice, the HFD-fed Plcb1f/f; Pdx1-CreERt2 mice also developed severe glucose intolerance due to a deficiency in insulin secretion. In addition, we confirmed that reduced insulin secretion was recovered in adding PLCβ1 back in MIN6 cell lines (Figure 2.15A, B). These result strongly support that effect on reduced insulin secretion was due to the PLCβ1 deletion and PLCβ1 can be considered a candidate to improve therapeutic intervention related to diabetes mellitus [55].
38 .
Figure 2.1. Expression profile of PLCβ isozymes in pancreatic β-cells
(A) Western blot analysis showing PLCβ isozyme proteins from extracts of MIN6 cells and isolated islets from wild type mice (C57BL6/J, 8 weeks old). (B, C) qRT-PCR analysis showing relative PLCβ isozyme mRNA levels in total RNA from MIN6 cells and islets isolated from wild type mice.
Lysates from the brain and thymus were used as positive controls.
39
Figure 2.2. Inducible ablation of PLCβ isozymes in adult pancreatic β-cells
(A) Graphical diagram of adult Plcb1, 2, 3, or 4f/f mice carrying the Pdx1-creERt2 transgene injected with tamoxifen to induce β-cell-specific deletion of PLCβ1, 2, 3, or 4 isozymes. (B) Western blot analysis showing the reduction in PLCβ isozyme protein contents from extracts of isolated islets from each Plcbf/f and Plcbf/f; Pdx1-creERt2 mice following tamoxifen treatment.
40
Figure 2.3. Impaired glucose tolerance in Plcb1f/f; Pdx1-CreERt2 mice
(A) Glucose tolerance test (2 mg glucose/g, intraperitoneal injection) for tamoxifen-treated Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice (n = 6 per group), (B) Plcb2f/f and Plcb2f/f; Pdx1-CreERt2 mice (Plcb2f/f, n = 12 vs. Plcb2f/f; Pdx1-CreERt2, n = 8), (C) Plcb3f/f and Plcb3f/f; Pdx1-CreERt2 mice (Plcb3f/f, n = 7 vs. Plcb3f/f; Pdx1-CreERt2, n = 10), and (D) Plcb4f/f and Plcb4f/f; Pdx1-CreERt2 mice (n = 5 per group). Error bars represent ± SEM. **p < 0.01 and *p < 0.05.
41
Figure 2.4. Defect in GSIS in Plcb1f/f; Pdx1-CreERt2 mice
(A) Plasma insulin and (B) glucose levels after overnight fasting (16 h), 15 min, and 30 min post glucose injection in Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice (Plcb1f/f, n = 5 vs Plcb1f/f; Pdx1- CreERt2, n=7). (C) Secreted insulin levels from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 islets. The results represent five independent experiments with islets from three mice per genotype. (D) Intracellular Ca2+ influx in isolated mouse islets from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice.
Pseudo-colored image of the fura-2AM fluorescence ratio (F340/F380) following high glucose treatment (left panel), the change in F340/F380 (middle panel, n = 11 per group), and the maximum change in F340/F380 (right panel, n = 11 per group). Error bars represent ± SEM. ***p < 0.001 and
**p < 0.01.
42
Figure 2.5. Normal insulin production and islet morphology
(A) Pancreatic sections stained with insulin and PLCβ1 antibodies. (B) mRNA expression of insulin genes from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 islets (three independent experiments with pooled islets from three mice per group). (C) Total insulin content of the pancreas (n = 4 per group). (D) Pancreatic sections from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice stained with H&E. (E) Insulin staining of pancreatic sections and quantification of β-cell area in Plcb1f/f and Plcb1f/f; Pdx1- CreERt2 mice (four to five islets were analyzed from the Plcb1f/f, n = 4 vs. Plcb1f/f; Pdx1-CreERt2, n = 5). Scale bars, low magnification: 200 µm, high magnification: 50 µm. Error bars represent ± SEM. *p < 0.05.
43
Figure 2.6. Decreased selective GqPCR-induced insulin release in Plcb1f/f; Pdx1-CreERt2 islets (A) Selective augmentation of insulin secretion in pancreatic islets. Isolated pancreatic islets prepared from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 were incubated for 1 h at 37°C in Krebs solution containing 16.7 mM glucose. In addition, the incubation medium contained one of the following ligands: AVP (5 μM), 5-HT (50 μM), KP (1 μM), CCK (500 nM), OXO-M (20 μM), and OA (10 μM). Supernatants were analyzed by ELISA. Data shown are representative of five independent experiments with islets from six to eight mice per genotype. (B) Pseudo-colored image of the fura- 2AM fluorescence ratio (F340/F380, upper panel), the change in F340/F380 (left panel), and the maximum change in the F340/F380 response (right panel) to 5-HT treatment (50 μM, n = 20 per group) and (C) to OXO-M treatment (20 μM, n = 25 per group). Blood glucose and insulin levels of Plcb1f/f; Pdx1-CreERt2 mice and Plcb1f/f littermates after receiving (D) a single dose of 5-HT (0.3 mg per mouse, Plcb1f/f, n = 5 vs. Plcb1f/f; Pdx1-CreERt2, n = 6) and (E) a single dose of bethanechol (2 µg/g, intraperitoneal injection, Plcb1f/f, n = 6 vs. Plcb1f/f; Pdx1-CreERt2, n = 7).
Plasma glucose and insulin levels were measured at the indicated time points. Scale bar: 400 µm.
Error bars represent ± SEM. ***p < 0.001, **p < 0.01, and *p < 0.05.
44
Figure 2.7. Impaired insulin secretion in Plcb1f/f; Pdx1-CreERt2 mice fed a HFD
(A) Body weight during HFD feeding (Plcb1f/f, n = 5 vs. Plcb1f/f; Pdx1-CreERt2, n = 4). (B) Pancreatic sections from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice stained with H&E and insulin staining of pancreatic sections (20 weeks old mice in figure 7A). (C) Glucose tolerance test (2 mg glucose/g, intraperitoneal injection, n = 7 per group). (D) Total insulin content of pancreas (n = 4 per group). (E) Plasma glucose and insulin levels from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice (n
= 5–7 per group) post glucose injection. (F) Secreted insulin levels from Plcb1f/f and Plcb1f/f; Pdx1- CreERt2 islets (three independent experiments with pooled islets from three mice per group). Scale bars, low magnification: 200 µm, high magnification: 50 µm. Error bars represent ± SEM. ***p <
0.001, **p < 0.01, and *p < 0.05.
45
Figure 2.8. Similar body weight and fasting and re-feeding glucose levels between control and PLCβ cKO mice following tamoxifen treatment
(A) Body weight of Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice (Plcb1f/f, n = 7 vs. Plcb1f/f; Pdx1-CreERt2, n = 8), Plcb2f/f and Plcb2f/f; Pdx1-CreERt2 mice (Plcb2f/f, n = 12 vs.
Plcb2f/f; Pdx1-CreERt2, n = 9), Plcb3f/f and Plcb3f/f; Pdx1-CreERt2 mice (Plcb3f/f, n = 6 vs. Plcb3f/f; Pdx1-CreERt2, n = 7), and Plcb4f/f and Plcb4f/f; Pdx1-CreERt2 mice (Plcb4f/f, n = 6 vs. Plcb4f/f; Pdx1-CreERt2, n = 5) mice. (B) Fasting and re-feeding glucose levels of Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice (Plcb1f/f, n = 7 vs. Plcb1f/f; Pdx1-CreERt2, n = 8), Plcb2f/f and Plcb2f/f; Pdx1-CreERt2 mice (Plcb2f/f, n = 8 vs. Plcb2f/f; Pdx1-CreERt2, n
= 9), Plcb3f/f and Plcb3f/f; Pdx1-CreERt2 mice (Plcb3f/f, n = 7 vs. Plcb3f/f; Pdx1-CreERt2, n = 8), and Plcb4f/f and Plcb4f/f; Pdx1-CreERt2 mice (Plcb4f/f, n = 6 vs. Plcb4f/f; Pdx1- CreERt2, n = 5) mice before and after tamoxifen.
46
Figure 2.9. Normal insulin sensitivity in Plcb1f/f; Pdx1-CreERt2mice fed a normal chow diet and HFD
(A) Insulin tolerance test (0.3 U/kg) for control and Plcb1f/f; Pdx1-CreERt2 mice fed a normal chow diet (Plcb1f/f, n = 6 vs. Plcb1f/f; Pdx1-CreERt2, n = 7) and (B) Plcb1f/f and Plcb1f/f; Pdx1- CreERt2 mice fed a HFD (n = 6 per group).
47
Figure 2.10. PLCβ1 protein expression in α-cell and δ-cells in pancreatic islets from control and Plcb1f/f; Pdx1-CreERt2 islets
Pancreatic sections stained with PLCβ1, glucagon and somatostatin antibodies. Scale bars, 50 µm.
48
Figure 2.11. Reduced PLCβ1 expression in human β-cells from type 2 diabetes patients compared to normal subjects
PLCβ1 expression in human β-cells from normal and type 2 diabetes patients (DM) plotted as relative units (Gene Expression Omnibus public repository, accession no. GSE20966).
49
Figure 2.12. Reduced PLCβ1 expression in hypothalamus results in no alteration in food intake
(A) Brain sections stained with PLCβ1 antibodies in hypothalamus. (B) Absolute food intake (n = 6 per group). 3V; Third ventricle. Scale bars, low magnification: 50 µm, high magnification: 100 µm. Error bars represent ± SEM.
50
Figure 2.13. No change in β-cell proliferation and β-cell turnover in Plcb1f/f; Pdx1-CreERt2 islets
(A) Pancreatic sections stained with insulin and Ki-67 antibodies. (B) Western blot analysis showing apoptotic protein contents from extracts of isolated islets from Plcb1f/f and Plcb1f/f; Pdx1-CreERt2 mice. Scale bars, 50 µm.
51
Figure 2.14. PLCβ isozyme expression in each PLCβ isozyme conditional knockout islets (A) Western blot and (B) qRT-PCR analysis showing all PLCβ isozyme contents from extracts of isolated islets from each PLCβ isozyme conditional knockout mice following tamoxifen treatment.
Error bars represent ± SEM. ***p < 0.001.
52
Figure 2.15. Reconstitution of PLCβ1 expression to rescue cells from abnormal insulin secretion
(A) Western blot analysis showing the PLCβ1 protein contents and (B) Secreted insulin level from PLCβ1-rescued MIN6 cell lines under high (16.7mM) glucose condition (n = 3). Error bars represent ± SEM. **p < 0.01 and *p < 0.05.
53
Gene Sequence (5’-3’)
Plcb1-F AGA TCA GCG AGG ACA GCA AT
Plcb1-R GCC CAG GCA GTG ATA TTT GT
Plcb3-F TTC GCC CTG ATG AGT TTC CC
Plcb3-R AGC ACT TCG TTG AGT CTC GG
Plcb4-F CCA CCG ACA CCA TAC GGA AA
Plcb4-R GGA GAT GTG TCG GTA GCC T
Ins-1-F GAC CAG CTA TAA TCA GAG ACC ATC
Ins-1-R GTA GGA AGT GCA CCA ACA GG
Ins-2-F GGC TTC TTC TAC ACA CCC AT
Ins-2-R CCA AGG TCT GAA GGT CAC CT
36b4-F TGG CCA ATA AGG TGC CAG CTG CTG
36b4-R CTT GTC TCC AGT CTT TAT CAG CTG CAC
Table 2.1. Sequences of quantitative RT-PCR primer upstream and downstream
54